Bone marrow stromal dysfunction in mice administered cytosine arabinoside

Objective. The aim of the study was to investigate ex-vivo the bone marrow (BM) stroma of mice under conditions of low- and high-dose cytosine arabino side (Ara-C), a cycle-specific drug (S-phase) and to assess possible stroma l damage, apart from the killing of hematopoietic cells. Stroma consists of mesenchymal elements generally not in the cell cycle; therefore it. could not be a target for the killing effect of Ara- C. Materials and Methods. Th e stromal function was studied by tile following: the incidence of stromal stem cells, i.e. CFU-F; formation of stromal layers under growth conditions of long-term culture (LTC) followed by irradiation and overlayering of tes t cells in contact and noncontact co-cultures; subsequent culture of the te st cells in a semi-solid medium to assay the incidence of hyperproliferativ e potential cells (HPPC); production of GM-CSF, IL-3, IL-4, IL-6 and IFN ga mma in the conditioned medium (CM) of confluent stromal layers. All tests a nd assays were carried out on BM specimens, 1-4 d after Ara-C administratio n and on controls. Results. Low-dose Ara-C induces a marked decrease of CFU -F, compensated by cycle induction of pre-CFU-F, young-type stromal stem ce lls. High-dose Ara-C causes a CFU-F decrease to almost zero level. The time length to layer confluency is normal after low-dose Ara-C (similar to 10 d ) and prolonged after a high dose (similar to 30 d). The confluent layers f rom mice receiving low- or high-dose Ara-C support hematopoiesis adequately . Among the growth factors and: cytokines assayed, only IL-6 is detected in CM layers. IL-6 decreases after a low dose of Ara-C and increases after a high dose. The cause of IL-6 fluctuations is yet to be investigated. It is, however, evident that IL-6 is not an essential factor in support of hemato poiesis. Conclusions. Taken together, the current study in mice indicates t hat Ara-C administration, in particular a high dose, induces bone marrow st romal damage and/or disfunction. The long period of time to roach layer con fluency after a high Ara-C dose might reflect the in-vivo situation of slow stromal regeneration.